Semantic Segmentation using Neural Ordinary Differential Equations

TL;DR

This paper introduces a neural ODE-based model for semantic segmentation that achieves state-of-the-art results while significantly reducing memory usage and model size across multiple datasets.

Contribution

The paper presents a novel neural ODE architecture for semantic segmentation, improving efficiency and performance over traditional residual networks.

Findings

Achieves state-of-the-art results on multiple datasets.

Uses 57% less memory for training and 42% less for testing.

Reduces the number of parameters by 68%.

Abstract

The idea of neural Ordinary Differential Equations (ODE) is to approximate the derivative of a function (data model) instead of the function itself. In residual networks, instead of having a discrete sequence of hidden layers, the derivative of the continuous dynamics of hidden state can be parameterized by an ODE. It has been shown that this type of neural network is able to produce the same results as an equivalent residual network for image classification. In this paper, we design a novel neural ODE for the semantic segmentation task. We start by a baseline network that consists of residual modules, then we use the modules to build our neural ODE network. We show that our neural ODE is able to achieve the state-of-the-art results using 57% less memory for training, 42% less memory for testing, and 68% less number of parameters. We evaluate our model on the Cityscapes, CamVid, LIP, and PASCAL-Context datasets.